SA Agents — AI-Powered Sales Engineering Team

A multi-agent system designed for Solution Architects (SAs) to streamline enterprise sales processes. These agents handle the full deal lifecycle from customer research to RFP submission.

Overview

SA Agents is a suite of three specialized AI agents that work together to support enterprise sales deals:

Agent	Name	Phase	Role
🎯 sa-solution-agent	Claw-Solution	20% CRA	Customer research, solution design, proposal creation
🔬 sa-poc-agent	Claw-POC	60% POC	Proof of concept planning, test cases, execution
📋 sa-rfp-agent	Claw-RFP	80% RFP	Tender analysis, bid response, compliance

Agent Capabilities

🎯 SA-Solution Agent (Sol)

The AI Sales Engineer

Handles the Customer Research & Analysis (CRA) phase:

• Customer needs analysis
• Competitive strategy development
• Decision chain mapping
• Solution design
• Technical proposal writing
• Quotation generation
• Win rate analysis

Trigger phrases:

• "needs analysis" — Research customer environment and pain points
• "competitive strategy" — Analyze competitors and positioning
• "decision chain" — Map decision makers and influencers
• "solution" — Design technical solutions
• "technical proposal" — Create formal proposals
• "quotation" — Generate pricing documents

---

🔬 SA-POC Agent (Claw-POC)

The AI Technical Proof Engineer

Handles the Proof of Concept (POC) phase:

• POC competitive strategy
• POC planning and tracking
• Test case design
• Daily progress logging
• POC report generation
• Post-POC review

Trigger phrases:

• "POC strategy" — Technical competitive prep
• "POC plan" — Create/update POC plan
• "test cases" — Build test case library
• "daily update" — Log POC progress
• "POC report" — Generate final report
• "POC review" — Prepare review materials

---

📋 SA-RFP Agent (Claw-RFP)

The AI Tender Strategist

Handles the Request for Proposal (RFP) phase:

• Technical control parameters strategy
• Tender analysis and compliance matrix
• Bid response drafting
• Pre-submission inspection

Trigger phrases:

• "control parameters" — Requirements influence strategy
• "tender analysis" — Analyze RFP requirements
• "bid response" — Write bidding documents
• "inspection" — Compliance check before submission

Workflow Sequence

┌─────────────────────────────────────────────────────────────────────────┐
│                         ENTERPRISE SALES FLOW                          │
└─────────────────────────────────────────────────────────────────────────┘

    20% CRA                    60% POC                   80% RFP
   ┌─────────┐               ┌─────────┐               ┌─────────┐
   │   Sol   │──────────────→│Claw-POC │──────────────→│Claw-RFP │
   │(Research│               │ (Proof  │               │ (Tender │
   │  + Plan)│               │ of Work)│               │ Response│
   └─────────┘               └─────────┘               └─────────┘
        │                         │                         │
        ▼                         ▼                         ▼
   • Needs Analysis          • POC Strategy          • Control Params
   • Competitive Strategy    • Test Cases            • Tender Analysis
   • Decision Chain          • Daily Tracking        • Bid Response
   • Solution Design         • POC Report            • Inspection
   • Technical Proposal
   • Quotation

Directory Structure

sa-agents/
├── README.md                     # This file
├── .gitignore
│
├── sa-solution-agent/            # 🎯 Sol — Solution Design Agent
│   ├── AGENTS.md                 # Main workspace hub with skills
│   ├── IDENTITY.md               # Agent identity and personality
│   ├── SOUL.md                   # Core operating principles
│   ├── TOOLS.md                  # Available tools and capabilities
│   ├── USER.md                   # User context template
│   ├── MEMORY.md                 # Long-term memory storage
│   └── memory/                   # Daily session logs
│
├── sa-poc-agent/                 # 🔬 Claw-POC — POC Agent
│   ├── AGENTS.md                 # Main workspace hub with skills
│   ├── IDENTITY.md               # Agent identity and personality
│   ├── SOUL.md                   # Core operating principles
│   ├── TOOLS.md                  # Available tools and capabilities
│   ├── USER.md                   # User context template
│   ├── MEMORY.md                 # Long-term memory storage
│   └── memory/                   # Daily session logs
│
└── sa-rfp-agent/                 # 📋 Claw-RFP — RFP Agent
    ├── AGENTS.md                 # Main workspace hub with skills
    ├── IDENTITY.md               # Agent identity and personality
    ├── SOUL.md                   # Core operating principles
    ├── TOOLS.md                  # Available tools and capabilities
    ├── USER.md                   # User context template
    ├── MEMORY.md                 # Long-term memory storage
    └── memory/                   # Daily session logs

Quick Start

1. Project Setup

Start with SA-Solution Agent to create a new project:

Trigger: "new project"
Output: Creates standardized project structure

2. Research Phase

Continue with SA-Solution Agent:

Trigger: "needs analysis" → Customer research
Trigger: "competitive strategy" → Competitor analysis
Trigger: "decision chain" → Org mapping

3. Solution Phase

Still with SA-Solution Agent:

Trigger: "solution" → Design solution
Trigger: "demo script" → Prepare demo
Trigger: "technical proposal" → Write proposal
Trigger: "quotation" → Generate pricing

4. POC Phase

Switch to SA-POC Agent:

Trigger: "POC plan" → Create POC plan
Trigger: "test cases" → Design tests
Trigger: "POC report" → Final report

5. RFP Phase

Switch to SA-RFP Agent:

Trigger: "tender analysis" → Analyze RFP
Trigger: "bid response" → Write bid
Trigger: "inspection" → Final check

Document Output Standards

All agents follow a consistent output format:

1. English Markdown (.md) — Primary customer-ready version
2. Chinese Markdown (.md) — Internal reference version
3. Export Format (.docx / .pptx / .xlsx) — Final formatted version

Project Structure Generated

{Project Name}/
├── 00_project-info.md
├── 20%-CRA/
│   ├── 01_Customer Needs Analysis/
│   ├── 02_Competitive Analysis/
│   ├── 03_Customer Decision Chain Analysis/
│   ├── 04_Solution Demonstration Script/
│   ├── 05_Solution/
│   ├── 06_Technical Proposal/
│   └── 07_Quotation/
├── 40%-CallHigh/
│   ├── 08_Senior Management Report/
│   └── 09_Project Win Rate Analysis/
├── 60%-POC/
│   ├── 10_POC Technical Competitive Strategy/
│   ├── 11_POC Planning and Tracking/
│   ├── 12_POC Test Cases/
│   ├── 13_POC Report/
│   └── 14_POC Review/
└── 80%-RFP/
    ├── 15_Technical Control Parameters/
    ├── 16_Tender Analysis Report/
    ├── 17_Bidding Documents/
    └── 18_Bidding Document Inspection/

Agent Memory System

Each agent maintains context across sessions:

• Daily Logs (memory/YYYY-MM-DD.md) — Session-by-session progress
• Long-term Memory (MEMORY.md) — Curated patterns and lessons
• Project Context (00_project-info.md) — Deal-specific information

Multi-Agent Coordination

Agents work together through shared files:

• SA-Solution Agent outputs feed into POC and RFP phases
• SA-POC Agent reports inform RFP bid responses
• Each agent "owns" specific directories and reads from others

Operational Guidelines

Green — Act Freely

• Reading project files and knowledge base
• Generating draft documents
• Creating and updating memory files

Yellow — Confirm Before Acting

• Sending documents externally
• Changing project scope
• Including pricing figures

Red — Never Without Approval

• Direct customer communication
• Production environment access
• Final bid submission

Requirements

These agents are designed to work with:

• Claude Code CLI
• Markdown-based document workflow
• Enterprise sales methodology (Cloudwise)

License

Internal use for Solution Architects and Sales Engineering teams.

---

Built for winning enterprise deals through systematic, AI-powered sales engineering.